Long-term exposure of the breast endogenous estradiol (E2) is widely to be an important in determining a women's risk of developing breast cancer. Previous studies have shown that breast fluid (BF) obtained from nipple aspirates has many times the concentration of estradiol in serum, and that the concentration is not reduced significantly after menopause when blood levels have declined. The purpose of the present study is to ascertain the factors and mechanisms by which estradiol levels are maintained in the breast and the biological activity associated with those levels. We will determine whether salivary estradiol, an estimate of diffusible estradiol, is more closely related to BF 32 than serum E2 in both pre- and postmenopausal women. We will determine whether adrenal androgen precursors can explain the maintenance of BF E2 in postmenopausal women. We will determine whether adrenal androgen precursors can explain the maintenance of BF E2 in postmenopausal women. We will determine whether BF E2 or blood levels of E2 are more closely associated with the concentration of an estrogen response gene product, pS2, and to the growth factor, EGF, in BF. From the ratios of steroid products in BF we will estimate the importance of the several potential precursors of BF E2 and the role of specific cytokines for biosynthesis of BF E2. Potential feedback effects of estrogen on the levels of BF 32 will be determined in women receiving ethinyl estradiol and in women taking the estrogen antagonist, tamoxifen. For these experiments we have developed specific and sensitive assays for salivary E2, BF E2, and BF EGF and a method for separating BF products in a highly efficient manner into phenolic (estrogen), neutral (androgen), and aqueous fractions so that many more substances can be measured from a small volume of BF. Additional assays will be developed according to established protocols. We have already collected samples from pre- menopausal women on a dietary intervention study, and we plan to recruit pre-menopausal women planning to start oral contraceptive use and women who will start tamoxifen treatment as well as a group of postmenopausal women for these studies. Being able to monitor the effect on BF E2 of factors involved in its biosynthesis and in its action in human subjects will give us insight into the ways in which one of the important determinants of breast cancer risk is regulated.